U.S. patent number 5,074,983 [Application Number 07/341,936] was granted by the patent office on 1991-12-24 for thin film testing method.
This patent grant is currently assigned to HMT Technology Corporation. Invention is credited to Atef H. Eltoukhy, Yassin Mehmandoust.
United States Patent |
5,074,983 |
Eltoukhy , et al. |
December 24, 1991 |
Thin film testing method
Abstract
A method of evaluating the start/stop lifetime of a thin-film
magnetic medium having an overcoat formed by sputtering a
carbon-containing overcoat on a magnetic-film substrate. Resistance
to erosion of the medium is evaluated by the time required to wear
away the overcoat during a high-speed abrasion treatment. Lubricity
is measured either in terms of the coefficient of friction of the
disc, or in terms or surface hydrophobicity, as gauged by one of a
number of measurable surface properties. The method allows for
rapid assessment of disc wear properties, and can be used to
optimize sputtering conditions required to achieve selected
hardness and lubricity properties in a disc overcoat.
Inventors: |
Eltoukhy; Atef H. (Saratoga,
CA), Mehmandoust; Yassin (Berkeley, CA) |
Assignee: |
HMT Technology Corporation
(Fremont, CA)
|
Family
ID: |
23339639 |
Appl.
No.: |
07/341,936 |
Filed: |
April 21, 1989 |
Current U.S.
Class: |
204/192.13; 73/7;
204/192.16; 73/150R |
Current CPC
Class: |
C23C
14/54 (20130101); G01N 3/56 (20130101); C23C
14/0605 (20130101) |
Current International
Class: |
C23C
14/54 (20060101); C23C 14/06 (20060101); G01N
3/56 (20060101); C23C 014/34 (); G01N 003/56 () |
Field of
Search: |
;73/7,15R
;204/192.15,192.16,192.2,192.13 |
References Cited
[Referenced By]
U.S. Patent Documents
Foreign Patent Documents
Primary Examiner: Nguyen; Nam X.
Attorney, Agent or Firm: Dehlinger; Peter J.
Claims
It is claimed:
1. A method of evaluating the expected start-stop lifetime of a
thin-film magnetic disc formed by sputtering a carbon-containing
overcoat on a thin-film substrate, comprising
(a) measuring the resistance to erosion of the overcoat by (a)
placing against the disc, a roller covered with an abrasive tape
and oriented to rotate with disc rotation, (b) rotating the disc,
and (c) with the roller pressed against the overcoat, measuring the
overcoat removal time required to remove a selected portion of the
overcoat;
(b) measuring the lubricity of the overcoat; and
(c) correlating the measured resistance to erosion and lubricity
with the resistance to erosion and lubricity measured in discs with
known start-stop lifetimes.
2. The method of claim 1, wherein the resistance to erosion is
measured by rotating the disc at a speed of about 1,800 rpm, under
conditions in which the hardest disc overcoats can be substantially
worn away within at most about 15 minutes.
3. The method of claim 1, wherein said lubricity measuring includes
measuring the coefficient of friction of a read/write head in
contact with disc.
4. The method of claim 3, wherein said lubricity measuring includes
measuring the coefficient of friction of a read/write head on the
disc, after a 30-minute head drag period at a disc speed of 300
rpm.
5. The method of claim 4, wherein the coefficient of friction is in
the range between about 0.2-0.5 before and after such head-drag
period.
6. The method of claim 1, wherein said lubricity measuring includes
measuring the hydrophobicity of the overcoat.
7. The method of claim 6, wherein said hydrophobicity measuring
includes measuring the angle from horizontal at which a bead of
water placed on the disc begins to glide.
8. The method of claim 6, wherein said hydrophobicity measuring
includes measuring the ratio of exposed carbon:oxygen-containing
groups on the overcoat surface.
9. The method of claim 8, wherein said ratio is greater than about
4:1.
10. The method of claim 6, wherein said hydrophobicity measuring
includes measuring the reflectivity of the disc.
11. The method of claim 6, wherein said hydrophobicity measuring
includes observing a bright gold color.
12. The method of claim 1, for use in optimizing sputtering
conditions used in achieving a desired balance between hardness and
lubricity in the overcoat of a thin-film magnetic disc, in forming
a graphite overcoat on a substrate by a sputtering process carried
out under a low-pressure, inert-gas atmosphere in a closed chamber
with a sputtering power level being developed for initiating
sputtering, which further includes adjusting the composition of the
gas present at low pressure during sputtering to increase the
lubricity of the overcoat, and varying the sputtering voltage and
pressure in the chamber to achieve a desired degree of
hardness.
13. The method of claim 12, wherein the composition of the gas
present during sputtering contains an inert gas and between about
20-70 mole percent low molecular weight hydrocarbon, and said
adjusting includes varying the relative mole ratios of the inert
gas and hydrocarbon.
14. A method of establishing gas pressure and sputtering voltage
conditions for sputtering a graphite overcoat on a substrate to
produce a selected start-stop lifetime in a disc formed by said
sputtering, said method comprising
measuring resistance to erosion of the graphite overcoat by (a)
placing against the disc, a roller covered with an abrasive tape
and oriented by rotate with disc rotation, (b) rotating the disc,
and (c) with the roller pressed against the overcoat, measuring the
overcoat removal time required to remove a selected portion of the
overcoat, measuring the lubricity of the graphite overcoat, and
adjusting said sputtering conditions to produce a selected
resistance to erosion and lubricity corresponding to such selected
start-stop lifetime.
15. The method of claim 14, wherein the resistance to erosion is
measured by rotating the disc at a speed of about 1,800 rpm, under
conditions in which the hardest disc overcoats can be substantially
worn away within at most about 15 minutes.
16. The method of claim 14, wherein said lubricity measuring
includes measuring at least one of the following surface
characteristics: (a) the static coefficient of friction of a
read/write head in contact with disc, (b) the dynamic coefficient
of friction of a read/write head in contact with disc, (c) the
angle from horizontal at which a bead of water placed on the disc
begins to glide, (d) the ratio of exposed carbon-carbon to
carbon-oxygen-containing groups on the overcoat surface, (e) the
surface reflectivity of the disc, and (f), color of the disc.
17. The method of claim 14, which further includes adjusting the
sputtering voltage in a sputtering apparatus, in response to said
measuring of resistance to erosion, to produce a selected hardness
in the overcoat.
18. The method of claim 14, which further adjusting the pressure in
a sputtering apparatus, in response to said measuring of resistance
to erosion, to produce a selected hardness in the overcoat.
19. The method of claim 14, which further adjusting the percent of
hydrocarbon gas in an argon/hydrocarbon gas mixture in a sputtering
apparatus, in response to said measuring of lubricity, to produce a
selected lubricity in the overcoat.
20. The method of claim 19, which further adjusting the sputtering
voltage and chamber pressure in a sputtering apparatus, in response
to said measuring of resistance to erosion, to produce a selected
hardness and lubricity in the overcoat.
Description
1. FIELD OF THE INVENTION
The present invention relates to a method of testing wear-related
characteristics of a magnetic thin-film disc having a carbon
overcoat, and the use of the method for optimizing overcoat
properties.
2. REFERENCES
Craig, S., et al, Thin Solid Films, 97:345 (1982).
Kobayashi, K., et al., Thin Solid Films, 158:233 (1988).
Natarajan, V., et al., J Vac Sci Technology, A3(3):681 (1985).
Research Disclosure RD269061, K. Mason Publications, Ltd., England
(1986).
Tsai, H-C., et al., J Vac Sci Technol, A5(6):3287 (1987).
Yolamanchi, R. S., et al., Thin Solid Films, 164:103 (1988).
3. BACKGROUND OF THE INVENTION
Magnetic overcoats are commonly formed on substrates, such as
magnetic thin-film discs employed for recording data. The thickness
of the overcoat is typically between about 200-500 .ANG. and
preferably about 300 .ANG.. Greater thickness of the overcoat, and
thus greater distance between the thin-film magnetic storage layer
and the read/write head, tends to degrade disc resolution and
storage density.
The carbon overcoat functions to protect the underlying magnetic
layer from damage and wear caused by repeated contact between the
disc and the read-write used in accessing the disc. For this
reason, the graphite overcoat is ideally formed to have a high
degree of hardness or erosion-resistance.
In addition, the graphite overcoat is intended to provide
lubricating surface properties, to minimize drag on the head and
wear on the disc during prolonged head/disc contact. The overcoat
therefore ideally provides a low-friction surface. The lubricity of
a hard carbon overcoat on a disc may be enhanced by covering the
overcoat with a thin liquid layer of a stable fluid material, such
as a fluorocarbon. The optimum friction reduction may be achieved
with a liquid layer of fluorocarbon of about 15-30 .ANG.. With a
thicker liquid layer, the head may tend to "blot" the liquid layer,
producing greater drag between the head and disc with a resultant
reduction in the operating lifetime for the disc.
A variety of methods have been used heretofore for forming carbon
overcoats on a thin-film magnetic disc (Tsai). In one method, known
as RF plasma or glow discharge, an RF source is used to decompose
an hydrocarbon gas, producing a carbonaceous plasma whose carbon
particles are deposited on a thin-film substrate to form the carbon
overcoat (e.g., Natarajan; Yolamanchi; and Kobayashi). The RF
discharge method is relatively slow, and deposition rates and
plasma composition are somewhat difficult to control.
Another method which has been used for producing a carbon overcoat
involves carbon deposition by DC sputtering, typically DC magnetron
sputtering, in which the ionized gases are directed onto the target
by magnetic fields established in the sputtering device. Typically
in this method, a graphite substrate is sputtered onto a thin-layer
film substrate in a low-pressure argon gas until an overcoat of the
desired thickness is reached.
The resulting carbon overcoat has a predominantly graphitic
structure with "islands" of diamond-like crystalline clusters with
dimensions on the order of about 20 .ANG.. It is, of course, the
diamond-like clusters which impart the hardness properties to the
overlayer. Although the overcoat formed in this manner has adequate
hardness properties, it would be desirable to increase the
lubricity of the layer as well, particularly the lubricity of the
overcoat after initial wear. Experiments conducted in support of
the present invention indicate that carbon overcoats formed by DC
magnetron sputtering in a pure argon atmosphere tend to show a
substantial loss of lubricity as the overcoat is worn, in turn,
causing greater wear on the overcoat. As a result, mechanical
stress in the system wearing away of the disc overcoat are both
accelerated.
The need for increased lubricity is especially great in the inner
diameter region of the disc, where the fluorocarbon liquid coating
applied to the overcoat becomes depleted over time due to migration
of the liquid material under centrifugal effects, and particularly,
in the inner-diameter region which is dedicated to start-stop head
contact, where repeated contact with the head further depletes the
liquid layer.
Various modifications of DC sputtering for use in producing carbon
overcoats have been proposed. One modification, for example, is to
form the overcoat by DC sputtering of a carbon target in the
presence of methane or a mixture of argon and methane (RD 269061;
Craig). This approach has the potential for increasing the
lubricity of carbon overcoat films, as has been verified by
experiments conducted in support of the present invention. However,
experiments conducted in support of the present invention indicate
that merely adding hydrocarbon gas to the sputtering chamber
increases lubricity, but also reduces hardness, i.e., resistance to
wear.
Heretofore, the development of methods for producing durable,
high-lubricity overcoats for thin-film media has been hampered by
slow and/or inaccurate disc testing procedures. Disc hardness is
measured, according to the prior art, by a static or dynamic
(moving disc) scratch test in which the depth of scratch or
indentation (at a given scratch pressure applied to the disc),
provides a measure of hardness. This test is difficult to
quantitate, and does not necessarily correlate well with the
expected start/stop lifetime of a disc.
The lubricity properties of carbon-overcoat discs are generally
measured in terms of static or dynamic (rotating disc) coefficients
of friction. This is done by a standard drag test in which the drag
produced by contact of a read/write head with a disc is determined.
One important property of a disc which is required for good
long-term disc and drive performance is that the disc retain a
relatively low coefficient of friction after many start/stop cycles
or contacts with a read/write head. For example, a drive
manufacturer may require that the disc have an initial coefficient
of static friction no greater than 0.3, and a coefficient of static
friction of no greater than 0.6 after 20,000 start/stop cycles.
This disc specification indicates that the disc can tolerate at
least 20,000 start/stop cycles without showing high friction
characteristics which would interfere with read/write
operations.
In fact, the above start/stop lifetime test, in which the
coefficient of friction is measured before and after a large number
of start/stop cycles, has heretofore been the only reliable measure
of start/stop lifetime of a disc.
An obvious limitation of this test is the length of time required,
as well as the wear on the testing machinery. More importantly, the
results of the test do not indicate how changes in hardness and/or
lubricity, as measured by prior art methods, would lead to greater
durability, or more generally, how the hardness and lubricity
properties of an overcoat can be varied to produce desired
durability and low-friction properties in a disc.
4. SUMMARY OF THE INVENTION
It is therefore one general object of the invention to provide a
method of testing thin-film media for surface hardness and
lubricity characteristics which substantially overcomes limitations
associated with prior art testing methods.
It is a related object of the invention to provide such a method
which can be used in conjunction with carbon overcoat sputtering
methods to achieve desired hardness and lubricity characteristics
in an overcoat.
In one aspect, the invention includes a method of evaluating the
expected start-stop lifetime of a thin-film magnetic medium formed
by sputtering a carbon-containing overcoat on a thin-film
substrate. The method includes the steps of:
(a) measuring the resistance to erosion of the overcoat by (a)
placing against the disc, a roller covered with an abrasive tape
and oriented to rotate with disc rotation, (b) rotating the disc,
and (c) with the roller pressed against the overcoat, measuring the
overcoat removal time required to remove a selected portion of the
overcoat;
(b) measuring the lubricity of the overcoat; and
(c) correlating the measured resistance to erosion and lubricity
with the same parameters in discs with known start-stop
lifetimes.
In one preferred method, the resistance to erosion is measured by
rotating the disc at a speed of about 1,800 rpm, under conditions
in which the hardest disc overcoats can be substantially worn away
within at most about 15 minutes.
The lubricity is measured, in accordance with one embodiment of the
invention, by measuring the dynamic coefficient of friction before
and after a given period of head drag on an unlubricated
overcoat.
In another general embodiment, the lubricity is measured in terms
of the hydrophobicity of the disc. This property may be determined
from the angle from horizontal at which of bead of water placed on
the disc begins to glide on the disc. Alternatively hydrophobicity
can be measured from the ratio of exposed carbon:oxygen-containing
groups on the overcoat surface. In addition, high reflectivity
and/or a bright gold overcoat color may characterize an overcoat
having the desired hydrophobic properties.
In another aspect, the invention includes a method for use in
establishing optimal operating parameters for forming a graphite
overcoat on a thin-film magnetic substrate by DC magnetron
sputtering of a carbon substrate in a low-pressure
argon/hydrocarbon atmosphere. The method includes the steps of:
(a) measuring resistance to erosion of the graphite overcoat,
(b) measuring the lubricity of the graphite overcoat, and
(c) based on the measured values of hardness and lubricity,
adjusting the ratio of argon:hydrocarbon in the sputtering
atmosphere to selectively vary the lubricity of the overcoat, and
adjusting the sputtering voltage and gas pressure in the sputtering
chamber, to produce a desired degree of hardness at the selected
lubricity.
These and other objects and features of the invention will become
more fully apparent when the following detailed description of the
invention is read in conjunction with the accompanying
drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a sectional view of a thin-film magnetic disc having a
carbon overcoat whose properties are to be tested and optimized in
accordance with the present invention;
FIG. 2 is a schematic representation of a device for rapidly
measuring carbon overcoat resistance to erosion, in accordance with
the invention;
FIG. 3 is a side view, with components schematically represented,
of a device for measuring the dynamic coefficient of friction of a
carbon overcoat;
FIG. 4 illustrates a method for measuring bead angle on a carbon
overcoat for determining hydrophobicity;
FIG. 5 is a schematic view of sputtering apparatus used in
producing the disc shown in FIG. 1;
FIG. 6 is a block diagram of sequential steps which may be employed
in various combinations to produce, in a thin-film magnetic medium,
a carbon overcoat having desired hardness and lubricity properties;
and
FIGS. 7A and 7B are ESCA spectra of carbon overcoats formed in the
presence (7A) and absence (7B) of a hydrocarbon gas in an
argon-containing atmosphere in a sputtering apparatus of the type
shown in FIG. 5.
DETAILED DESCRIPTION OF THE INVENTION
The testing method of the invention is described in Section I
below. Section II describes a sputtering method which is used in
conjunction with the testing method to achieve desired overcoat
hardness and lubricity properties in a carbon overcoat. Properties
of a carbon overcoat formed in accordance with the testing and
sputtering method are detailed in Section III.
I. Testing Carbon Overcoat Surface Properties
FIG. 1 is a fragmentary portion of a thin-film medium or disc 10
constructed according to the method described in Section II, and
whose resistance to erosion, lubricity, and start/stop lifetime
characteristics are to be tested, in accordance with the present
invention.
The disk generally includes a rigid substrate 12, and forming
successive thin-film layers over the substrate, a crystalline
underlayer 14, a magnetic thin film layer 16, and a carbon overcoat
18, preferably about 200-400 .ANG.. The magnetic thin film layer
formed on the substrate is also referred to herein as a thin-film
substrate. It is understood that FIG. 1 illustrates only one of the
two sides of disk 10, the "lower" magnetic recording surface having
substantially the same construction as the upper surface which is
shown.
According to an important aspect of the invention, it has been
discovered the start/stop lifetime of a carbon overcoat on a disc
can be reliably estimated by combining resistance-to-erosion
measurements, described below, with a measured lubricity property
of the overcoat. The measured lubricity may be the static or
dynamic coefficient of friction of the overcoat, or one of a number
of properties related to the hydrophobicity of the overcoat, also
as detailed below.
FIG. 2 is a schematic side view of a device 20 used in measuring
the resistance to erosion of the carbon overcoat, in accordance
with the invention. The device used in the method is a commercially
available surface texturing machine conventionally used for
removing surface irregularities on a disc and/or for producing a
roughening or texturing a smooth disc surface, and whose operation
is modified as described below. Device 20 includes a rotor 22 on
which a disc, such as disc 24, is mounted for rotation about the
disc axis, indicated by dotted line 26. The rotor is driven for
rotation by a motor 28 whose speed can be adjusted for operation up
to about 1,800 rpm.
A pair of rollers 30, 32 in the device are mounted on a frame 34
for rotatable contacting the confronting surface of a disc during
operation of the device. Each roller, such as roller 30, is carried
on an axle, such as axle 36, for rotation about its long axis when
the roller is brought into contact with the spinning disc. Each
axle is mounted on a track 38 in the frame for sliding movement
thereon which moves the associated roller toward and away from the
confronting side of a disc. The two roller axles are connected by a
variable-tension spring 40, for applying a desired force of the
rollers against the opposite sides of the disc. One preferred
device of this type is commercially available form Exclusive Design
Company (San Mateo, Calif.), Model No. 800 HDF-C.
According to one modified aspect of the operation of the device,
the rollers are provided with an abrasive surface which is
effective to wear away the overcoat during operation in a period of
typically 5-15 minutes, at a selected force setting of the rollers,
and selected disc speed. One preferred abrasive surface is a
0.3.mu. particle size abrasive tape, such as tape No. 511904569
supplied by 3M Corp. (Minneapolis, Minn.).
The rollers are moved against the opposite disc surfaces with a
force of typically between about 3-4 pounds, by adjustment of
spring 40. The disc is then rotated at a selected speed, preferably
about 1,800 rpm. According to another modified aspect of the
operation of the device, the thickness of the overcoat is
monitored, e.g., by visual inspection, or by a chromometer, to
determine the time required to wear away a selected portion of the
overcoat. Preferably the operation of the device is adjusted so
that the overcoat is completely worn away in a period of several
minutes. For, example, using the settings and abrasive material
mentioned above, it has been found that a relatively soft carbon
overcoat is worn away in about 2 minutes, as compared with a hard
overcoat which is worn away in about 10 minutes.
As mentioned above, the resistance-to-erosion measurements above
are combined lubricity measurements to determine total start/stop
lifetime of the disc overcoat. One measure of lubricity is the
static or dynamic coefficient of friction of the overcoat. FIG. 3
is a schematic view of a device 42 used in measuring the dynamic
coefficient of friction of a carbon overcoat surface on a disc,
such as a disc 44. The device used in this test is a standard drag
test for measuring the static and dynamic coefficient of a disc,
such as the Model No. UTS-777 machine supplied by New Phase
Technology (San Jose, Calif.).
Briefly, the device includes a motor-driven rotor, indicated at 46,
for rotating the disc at a selected speed. A standard 9.5 g
minicomposite read/write head 48 in the device is mounted for
movement between positions of contact (solid lines) and non-contact
(dotted lines) with the disc. This head, in turn, is coupled to a
force transducer 50 which measures the force (in the direction of
disc movement) applied to the head upon contact with the disc. The
force measurement is used to determine the coefficient of friction.
The dynamic coefficient of friction is measured at 1 rpm.
In one preferred testing method, the dynamic coefficient is
measured, as above, both before and after a short drag period,
typically 30 minutes. Here the 9.5 g read/write head is dragged for
30 minutes on an unlubricated disc rotating at about 300 rpm.
Ideally, an overcoat will retain a relatively high lubricity (low
coefficient of friction) after the 30-minute drag period.
The lubricity of the disc overcoat can also be measured in terms of
the surface hydrophobicity of the overcoat. The correlation between
greater hydrophobicity and greater lubricity appears to be due, at
least in part, to the reduced degree of hydration which occurs on a
hydrophobic disc surface. That is, reduced hydration of polar
(hydrophilic) chemical groups, reduces the frictional drag at the
overcoat surface.
One simple measure of hydrophobicity is provided by a bead angle
test, such as illustrated in FIG. 4. Here a drop of water 50 is
placed on the upper surface of a disc, such as disc 52, and the
disc is raised above horizontal to an angle .alpha. at which the
bead first begins to glide on the surface, i.e., slide down the
surface. Bead angles of between about 50 and 70 degrees, and
preferably 60-70 degrees are typically observed. By comparison a
conventional carbon overcoat formed according to prior-art methods
has a bead angle of about 45 degrees or less and a lubricated
(highly hydrophobic) overcoat has a bead angle of 90 degrees.
Chemical analysis of the surface chemical groups on the overcoat
provides a more direct measure of hydrophobicity. One preferred
method of surface chemical analysis is Electron Spectroscopy for
Chemical Analysis (ESCA), which yields an electron density spectrum
over a defined range of electron binding energies, where the type
and density of characteristic chemical bonds on the surface can be
identified by the position and areas of the peaks in the spectrum,
respectively.
FIGS. 7A and 7B shows typical ESCA spectra for a carbon overcoat,
over a spectral energy region from about 278-298 eV. The peaks are
related to the 1s carbon electrons in C--C (284.38 eV), C--O
(285.97 eV), C.dbd.O (287.81 eV), and O--C.dbd.O (289.83 eV)
chemical groups, as indicated. As will be discussed below in
Section III, the hydrophobicity in a preferred carbon overcoat is
characterized by a ratio of C--C to C-oxygen containing bonds of at
least about 4:1.
Also as well be described below, a disc having desired
hydrophobicity properties is also characterized by a highly
reflective surface and a bright gold color. These optical
properties may therefore also be used to confirm that a disc has
the desired hydrophobic surface properties.
In practicing the method of the invention, the resistance to
erosion and lubricity of a disc are measured as detailed above. For
example, resistance to erosion and dynamic coefficient of friction,
and/or resistance to erosion and ESCA characteristics are measured.
The measured values are then correlated with the same values
determined for a surface overcoat with a known start/stop lifetime.
Known lifetimes are measured conventionally by a standard
start/stop test in which a read/write head is repeatedly brought
into contact with the surface of a spinning disc, until the
coefficient of friction of the disc increases above a selected
threshold, e.g., 0.6.
By way of example, the actual start/stop lifetimes of a series of
disc carbon overcoats having increasing resistance-to-erosion times
between 5-10 minutes and, for each resistance to erosion time,
having increasing coefficients of friction (after wear) of between
0.2-0.8 are determined. From these values, the expected lifetime of
a carbon overcoat having a measured resistance to erosion and
frictional coefficient can be determined.
It can be appreciated that the present invention allows the
expected start/stop lifetime of a disc to be measured rapidly and
with little wear on testing machinery. Thus, for example, a 10
minute resistance to erosion test, combined with measurements of
the dynamic coefficient of friction (before and after 30 minutes of
head drag) can now replace an extended test requiring several
thousand disc/head contact cycles.
In addition to its use in disc quality control, the method is
useful in evaluating the performance characteristics of an
overcoat, for purposes of establishing the optimal operating
parameters for forming a carbon overcoat in a sputtering method.
This application of the present invention will be described below
in Section III.
II. Sputtering Method
FIG. 5 is a schematic view of a sputtering apparatus 54 used in
forming a carbon overcoat having desired hardness and lubricity
characteristics. The apparatus generally includes a sputtering
chamber 56 with one or more sputtering stations at which a
thin-film layer is deposited on a substrate 58, typically a metal
disc substrate as shown. The substrate is carried through the
apparatus on a pallet 60, in a conveyor fashion, exposing each side
of the disc successively to the sputtered material in the
sputtering chambers.
Typically, the apparatus includes a first station at which a
crystalline underlayer 62 is deposited, a second station at which a
thin-film magnetic layer 64 is deposited, and a third station 66,
shown in the figure, at which the final carbon overcoat is
deposited. The sputtering conditions which are described below
apply to station 66 in particular.
Station 66 houses a pair of carbon, i.e., graphite targets, 68, 70.
The targets are connected to a power supply 72 in the apparatus to
achieve a selected target voltage with respect to the disc, as
shown. With a carbon target for use in forming a carbon overcoat,
the voltage is typically adjustable between about 500 to 600 volts,
giving a power level between about 0.8 and 4 kwatts.
The final pressure in the chamber during a sputtering operation is
a selected pressure preferably between about 10.sup.-3 to 10.sup.-2
mBarr. The vacuum pressure is achieved with a conventional
low-pressure vacuum pump 77.
A sputtering apparatus of the type just described is commercially
available, such as from Circuits Processing Apparatus (Fremont,
Calif.), Leybald Heraeus (Germany), VACTEK (Boulder, Colo.) or
Materials Research Corp (Albany, N.Y.). These systems are all
double-sided, in-line, high-throughput machines having two
interlocking chambers for loading and unloading.
The carbon overcoat on the substrate is formed by sputtering under
a low-pressure mixed-gas atmosphere. In accordance with the method,
this atmosphere is composed of a selected amount of low-molecular
weight hydrocarbon gas, preferably methane or ethane, and an inert
gas, preferably argon. The percentage of these gases in the chamber
is controlled by suitable valving of hydrocarbon and inert gas
containers 74, 76, respectively. The percentage (mole percentage)
of hydrocarbon gas in the chamber during overcoat formation is
preferably at least about 20 percent, and preferably no higher than
about 70 percent, with argon making up the remainder of the
gas.
FIG. 6 is a flow diagram showing how the testing method of the
invention is used in optimizing sputtering conditions, to produce a
carbon overcoat with selected hardness and lubricity properties.
The first step shown is a conventional sputtering step in which a
carbon overcoat is formed in a pure argon atmosphere, at a selected
vacuum pressure. The voltage (and resulting power level) of the
sputtering device is generally set close to the upper limits of the
apparatus, e.g., about 580-600 volts. The voltage and rate of
movement of the disc through the sputtering chamber are adjusted to
produce a final overcoat thickness of between about 200-500 .ANG.,
and preferably close to 300 .ANG..
The carbon overcoat formed by sputtering in pure argon, as above,
is tested for resistance to erosion and lubricity, as described in
Section I. Generally the overcoat has good resistance to erosion
but only moderate lubricity, as measured for example, by bead
angle.
A carbon overcoat is next formed by sputtering in a low-pressure
atmosphere containing a selected percentage of hydrocarbon gas,
e.g., from 10-70 mole percent methane. The addition of low
molecular weight hydrocarbon gas is initially made without any
adjustment in sputtering voltage or in the chamber pressure. The
carbon overcoat produced under these conditions shows increased
lubricity, as measured both by coefficient of friction and bead
angle, with increasing amounts of methane, up to about 30-50 mole
percent methane. However, the resistance to erosion time decreases
significantly with increasing methane concentrations.
According to an important feature of the sputtering method, it has
been discovered that with suitable adjustments in sputtering
voltage and gas pressure, both enhanced lubricity and high
resistance to erosion times can be obtained.
Considering the voltage and power levels parameter, it is observed
that voltage drops with increasing amounts of hydrocarbon gas in
the chamber, presumably due to the lower impedance of the (ionized)
hydrocarbon gas atmosphere. For example, a power supply setting
which gives a sputtering voltage of about 600 volts in pure argon
gives about 500 volts in 30-50% methane, at the same gas
pressure.
The resistance to erosion of a disc overcoat prepared in
methane:argon (50:50) at increasing voltage (and power levels) has
been examined. Briefly, it was discovered that at a voltage of
approximately 80-90% of the "pure argon" voltage--i.e., the optimal
level employed for sputtering in a pure inert gas--that greater
resistance to erosion was achieved, without loss of lubricity. The
adjusted increase in voltage produces a corresponding increase in
the power level for sputtering. Greater hardness (resistance to
erosion) can be achieved at higher voltages.
It has also been discovered that additional increase in resistance
to erosion can be achieved, also without loss of lubricity, by
decreasing the chamber pressure of the hydrocarbon-containing gas
mixture. It is noted here that this decrease in pressure tends to
increase sputtering voltage, so that this parameter needs to be
adjusted to an optimal level, as specified above. The final
pressure in the chamber is preferably about 40% that in the chamber
employing pure argon for sputtering. For example, the mixed gas
pressure may be about 2-3.times.10.sup.-3 mBarr. The resistance to
erosion can be increased or reduced by lowering or raising the
chamber pressure, respectively, from this level.
Once voltage, power level and gas pressure are adjusted for optimal
hardness, at a given gas composition, the percentage of hydrocarbon
gas may be further adjusted to enhance lubricity. By testing
resistance to erosion at the same time, the voltage, power level
and gas pressure conditions can also be systematically refined to
maintain or increase overcoat hardness, without loss of
lubricity.
It will be appreciated from above how the testing method of the
invention facilitates the selection of optimal sputtering
conditions, for achieving desired hardness and lubricity
properties. The resistance-to-erosion test, by providing a
quantitative measure of overcoat hardness, allows sputtering
conditions to be systematically controlled to produce a desired
hardness. At the same time, by selecting sputtering conditions
which also produce good lubricity, the method assures a disc with
good overcoat durability and long start/stop lifetime.
III. Overcoat Surface Properties
The testing and sputtering methods described have been employed to
produce a thin-film medium carbon overcoat having the following
thickness, and improved hardness, and lubricity
characteristics.
Thickness. The carbon overcoat has a thickness less than about 500
.ANG. and preferably about 300 .ANG., when used as a protective
coating in a thin-film magnetic recording medium. For other
applications, for example, where the overcoat is used to provide a
protective lubricating surface for plastic or metal mechanical
parts, the film thickness may be greater (or less) than this
specified thickness. The thickness of the overcoat may be
controlled, as above, by suitable adjustment in the time-of-travel
of the article to be coated through the carbon sputtering
station.
Resistance to Erosion. Resistance-to-erosion, as defined herein, is
measured in terms of the time required to wear away a 300 .ANG.
overcoat on a disc, when a 0.3.mu. abrasive tape is applied to the
surface of the disc with a force of about 3.5 pounds, and the disc
is rotated at 1,800 rpm. These test conditions, although
arbitrarily selected, are useful for making quantitative
comparisons of overcoat hardness which can be used both for
optimizing sputtering conditions and for determining start/stop
lifetimes, as above.
The carbon overcoat produced in accordance with above methods has a
resistance-to-erosion time of at least about 8 minutes, with
erosion times up to 10 minutes being achievable. An overcoat having
this degree of hardness, as well as high lubricity, is generally
well suited for conventional hard-disc computer storage
applications. However, applications which favor softer carbon
overcoats are also contemplated.
Coefficient of Friction. The coefficient of friction, as defined
herein, is the coefficient of friction measured with a standard
read/write head. The dynamic coefficient is measured at a disc
rotating at a speed of 1 rpm, and the static coefficient, from a
stationary disc position.
The carbon overcoat has a static and dynamic coefficient of between
about 0.2 and 0.3. To measure static and dynamic coefficient of
friction after extended start/cycles, a disc coated with
fluorocarbon lubricant was subjected to 30,000, 60,000, and 90,000
start/stop cycles, and tested for static coefficient. Even after
90,000 start/stop cycles, the static coefficient, as measured in
several disc, was between 0.2 to 0.45.
Similarly, the disc showed little or not increase in dynamic
coefficient of friction after a 30-minute head drag period, at a
disc speed of 300 rpm.
By contrast, a disc overcoat formed in accordance with prior art
sputtering methods showed gave static coefficient of friction
values of to 0.6-0.7 after 20,000 start/stop cycles.
Hydrophobicity: Bead Test. The bead angle at which a bead (e.g., a
10-20 .mu.l bead) of distilled water first begins to glide on the
overcoat surface is between about 50 and 70 degrees and preferably
between 60-70 degrees, as measured from the horizontal.
Hydrophobicity: Carbon Bond Ratio. FIGS. 7A and 7B are plots of the
ESCA spectra of carbon overcoats prepared in the presence (7A) and
absence (7B) of hydrocarbon gas in the sputtering chamber. The
spectra were broken down into peak components by a standard curve
fitting program, and the calculated peaks, representing C--C bonds
(284.38 eV), alcohol C--O (285.97 eV), ketone C.dbd.O (287.81 eV)
and acid and aldehyde O--C.dbd.O groups (289.83 eV), are shown in
dotted lines in the two figures. The percent of each bond type on
the surface, calculated by area under the curve, is given in Table
I for an overcoat formed in either pure argon atmosphere (third
column) and in a methane/argon atmosphere (fourth column).
TABLE I ______________________________________ Percent Percent
Energy Bond (argon) (argon/methane)
______________________________________ 284.58 C--C 71.5 81.7 285.79
C--O 14.8 12.9 287.42 C.dbd.O 8.7 4.2 289.56 O--C.dbd.O 5.0 1.2
______________________________________
As seen, the overcoat formed in an argon/hydrocarbon gas has a
ratio of carbon-carbon (hydrophobic) bonds to carbon-oxygen
(hydrophilic) bonds of greater than 4:1, and substantially higher
than an overcoat produced by carbon sputtering in a pure inert gas
environment.
Optical Properties. The carbon overcoat formed in accordance with
the sputtering and test methods described herein is also
characterized by high reflectivity. Although this property has not
been quantitated, initial visual observations indicate that the
disc overcoat is substantially more reflective than carbon
overcoats in commercially available discs.
The present carbon overcoat also has a distinctive bright gold
color. This color is easily distinguished from the typically grey
to brown color of known disc overcoats. It is assumed that the high
reflectivity and/or bright gold color of the overcoat are related
to the hydrophobicity of the disc and therefore may be diagnostic
of a disc overcoat having desired hydrophobicity
characteristics.
Start-Stop Lifetime. As defined herein, "start/stop lifetime" is
defined as the number of start/stop cycles which a lubricated disc
overcoat may be subjected to without an increase in static or
dynamic coefficient, beyond an acceptable specified value, e.g.,
0.5-0.7. The start/stop lifetime is measured conventionally be
measuring static coefficient of a lubricated disc before and after
an increasing number of start/stop cycles, e.g., 20,000 and 10,000
increments thereof.
By such standard start/stop lifetime testing, the present overcoat
(coated with a fluorocarbon lubricant) gave up to 90,000 start/stop
cycles without significant increase in static coefficient, as
mentioned above. This compares with the overcoat wear properties of
commercially available discs which show significant increases in
static coefficient of friction after 20,000-30,000 start/stop
cycles.
Although the invention has been described with respect to preferred
test methods, and their application to producing thin-film disc
carbon overcoats by DC magnetron sputtering, it will be appreciated
that various changes and modifications may be made without
departing from the invention .
* * * * *